JP2014060531A - Receiver - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate or reduce noise mixed in a double side band.SOLUTION: A receiver 1 comprises: a detection section 20 that demodulates a double side band amplitude modulation signal and extracts a first signal for an upper side band and a second signal for a lower side band; a band division section 50 that divides the first signal and the second signal for each bandwidth to generate a plurality of band signals; in-phase component extraction sections 60-1 to 61-n that extract in-phase component signals for each bandwidth between the first and second signals from the plurality of band signals; a first uncorrelated signal extraction section 80U that extracts a first uncorrelated signal from the first signal and each in-phase component signal; a second uncorrelated signal extraction section 80L that extracts a second uncorrelated signal from the second signal and each in-phase component signal; a gain adjustment section 90 that adjusts levels of each in-phase component signal with the first uncorrelated signal and the second uncorrelated signal; and a combining section 100 that adds each in-phase component signal with the first uncorrelated signal and the second uncorrelated signal after the level adjustment and generates an output signal.

Description

  The present invention relates to a receiving apparatus.

  Conventionally, an amplitude modulation signal including a double sideband (double sideband amplitude modulation signal) is received, and either the upper sideband or the lower sideband is extracted from the double sideband amplitude modulation signal, and the extracted sidebands are extracted. A receiver that demodulates a band is known (for example, see Patent Document 1). Noise mixed in a frequency region higher than the carrier frequency (carrier frequency) affects the upper sideband, and noise mixed in a frequency region lower than the carrier frequency affects the lower sideband. Therefore, the above receiver can acquire a baseband signal that is not affected by noise by selecting and using a sideband opposite to the frequency region where interference occurs, with the carrier frequency as the boundary. it can.

Japanese Patent No. 4440255

However, when interference occurs in both the upper sideband and the lower sideband, the receiver according to the above method cannot remove the noise.
The present invention has been made to solve the above problem, and an object of the present invention is to provide a receiving apparatus that removes or reduces noise mixed in both sidebands.

  [1] In order to solve the above-described problem, a receiving apparatus according to an aspect of the present invention demodulates a received double sideband amplitude modulation signal, and a first signal for the upper sideband and a second signal for the lower sideband. A detection unit that extracts a signal, a band division unit that generates a plurality of band signals by dividing each of the first signal and the second signal extracted by the detection unit for each band, and the band division unit generates An in-phase component extraction unit that extracts a signal of an in-phase component for each of the bands between the first signal and the second signal, and each of the first signal and the in-phase component extraction unit extracted from the plurality of band signals A non-correlated signal extracting unit that extracts a difference between the in-phase component signal as a first uncorrelated signal and extracts a difference between the second signal and the in-phase component signal as a second uncorrelated signal; In-phase component signal and A gain adjusting unit that adjusts the levels of the first uncorrelated signal and the second uncorrelated signal extracted by the Seki signal extracting unit, and a signal of each in-phase component that is adjusted by the gain adjusting unit, A combining unit that adds the uncorrelated signal and the second uncorrelated signal to generate an output signal.

[2] In the receiving device according to [1], the in-phase component extraction unit performs adaptive filter processing based on a learning identification method for each band, thereby obtaining the first signal and the plurality of band signals from the plurality of band signals. The in-phase component signal for each band between the second signals is extracted.
[3] The receiving apparatus according to any one of [1] and [2], further including a noise analysis unit that analyzes a noise component based on a difference between the first signal and the second signal extracted by the detection unit. The gain adjustment unit adjusts the level based on the analysis result of the noise component by the noise analysis unit.
[4] In the receiving device according to [3] above, a storage unit that stores a gain setting value of the gain adjustment unit according to the analysis result of the noise component by the noise analysis unit, and the gain setting from the storage unit A control unit that reads a value and sets the gain setting value in the gain adjustment unit, and the gain adjustment unit adjusts the level according to the gain setting value set by the control unit. It is characterized by.
[5] The receiving apparatus according to [3], further including: a filter unit that performs a filtering process on the output signal generated by the synthesis unit based on the analysis result of the noise component by the noise analysis unit. Features.

  According to the receiving apparatus of the present invention, noise mixed in both sidebands can be removed or reduced.

It is a figure which shows typically the spectrum of the signal strength with respect to the frequency of the double-sideband amplitude modulation signal obtained by modulating with an audio signal and the noise component mixed in the double-sideband amplitude modulation signal. It is a block diagram which shows the function structure of the receiver which is one Embodiment of this invention. It is a block diagram which shows the function structure of the in-phase component extraction part in the embodiment.

  Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. One embodiment of the present invention receives an amplitude modulation signal (double sideband amplitude modulation signal) including double sidebands (upper sideband and lower sideband) and demodulates the double sideband amplitude modulation signal. The receiving apparatus extracts the first signal for the upper sideband and the second signal for the lower sideband, and removes or reduces the noise component based on the first signal and the second signal. In the present embodiment, the double-sideband amplitude modulation signal is a signal obtained by amplitude-modulating a carrier wave with an audio signal by a conventional modulation device.

First, signals received by the receiving apparatus according to the present embodiment will be described. The signal received by the receiving apparatus according to the present embodiment is a double-sideband amplitude modulation signal and a noise component mixed in the double-sideband amplitude modulation signal. The noise component is generated by mixing of a noise signal due to an external interference wave, mixing of a signal from an adjacent frequency band, or the like.
FIG. 1 is a diagram schematically showing a spectrum of signal intensity with respect to frequency of a double sideband amplitude modulation signal obtained by modulation with an audio signal and a noise component mixed in the double sideband amplitude modulation signal. . As shown in the figure, the frequency spectrum of the double-sideband amplitude modulation signal includes a carrier signal and an upper-band signal that is higher than the carrier frequency by the frequency of the baseband signal (for example, an audio signal). , Each frequency spectrum with a lower sideband signal which is a signal lower than the frequency of the carrier wave by the frequency of the baseband signal. Further, the frequency spectrum of the upper side noise component (first noise component) exists in a frequency region higher than the carrier wave. In addition, a frequency spectrum of a lower side noise component (second noise component) exists in a frequency region lower than the carrier wave. The upper side noise component and the lower side noise component may be mixed independently of each other. Therefore, the frequency band and noise intensity of the upper side noise component and the lower side noise component may be different from each other. The figure shows that the upper-side noise component and the lower-side noise component are independent from each other, and the upper-side noise component and the lower-side noise component have different intensities, and the upper-side noise component is in the upper-side band. It shows a state in which the lower sideband noise component is partially mixed in the lower sideband.

Next, the configuration of the receiving apparatus according to the present embodiment will be described.
FIG. 2 is a block diagram illustrating a functional configuration of the receiving apparatus according to the present embodiment. As shown in the figure, the receiving apparatus 1 includes a control unit 10, a detection unit 20, a noise analysis unit 30, a storage unit 40, a band division unit 50, an in-phase component extraction unit 60, and a first signal delay. Unit 70U, second signal delay unit 70L, first uncorrelated signal extraction unit 80U, second uncorrelated signal extraction unit 80L, gain adjustment unit 90, synthesis unit 100, and filter unit 110. However, the receiving device 1 includes a plurality of in-phase component extraction units 60. Therefore, in the present embodiment, the plurality of in-phase component extraction units 60 are described as in-phase component extraction units 60-1 to 60-n (n is an integer of 2 or more, and the same applies hereinafter). The plurality of in-phase component extraction units 60 may be collectively referred to as an in-phase component extraction unit.

  The gain adjusting unit 90 includes an in-phase component gain adjusting unit 91, a first uncorrelated signal gain adjusting unit 92U, and a second uncorrelated signal gain adjusting unit 92L. However, the gain adjustment unit 90 includes a plurality of in-phase component gain adjustment units 91. Therefore, in the present embodiment, the plurality of in-phase component gain adjustment units 91 are referred to as in-phase component gain adjustment units 91-1 to 91-n. The plurality of in-phase component gain adjustment units 91 may be collectively referred to as an in-phase component gain adjustment unit.

  Further, the first uncorrelated signal extracting unit 80U and the second uncorrelated signal extracting unit 80L may be collectively referred to as an uncorrelated signal extracting unit.

The detection unit 20 takes in a double sideband amplitude modulation signal obtained by receiving the broadcast double sideband amplitude modulation wave with a receiving antenna (not shown), demodulates the double sideband amplitude modulation signal, A first signal for the band and a second signal for the lower sideband are extracted. Assume that the first signal and the second signal contain noise components. That is, a signal in phase (in-phase component signal) between the first signal and the second signal is C (bold body), and uncorrelated signals (first uncorrelated signal and second uncorrelated signal) are U (bold body). ₀ , L (bold body) Assuming ₀ , the first signal U (bold body) and the second signal L (bold body) are expressed by the following equation (1). Note that the description “(bold)” indicates that the character immediately before it is bold, and that the character is a vector.

When a part of the noise component is superimposed on the double sideband in the double sideband amplitude modulation signal, the in-phase component signal C includes a baseband signal and a part of the noise component. The first uncorrelated signal U (bold body) ₀ and the second uncorrelated signal L (bold body) ₀ include a portion excluding the part of the noise component.

  The detection unit 20 supplies the first signal to the band division unit 50, the noise analysis unit 30, and the first signal delay unit 70U, and supplies the second signal to the band division unit 50, the noise analysis unit 30, and the second signal delay. Part 70L.

  The noise analysis unit 30 takes in the first signal and the second signal supplied from the detection unit 20, and analyzes the noise component based on the first signal and the second signal. Then, the noise analysis unit 30 determines a gain setting value for the gain adjustment unit 90 and a filter coefficient for the filter unit 110 based on the analysis result, and stores the gain setting value and the filter coefficient in the storage unit 40. . The gain setting value is a gain (gain) for each of the in-phase component gain adjusting units 91-1 to 91-n, the first uncorrelated signal gain adjusting unit 92U, and the second uncorrelated signal gain adjusting unit 92L. This is setting information for each gain adjustment unit for determination. That is, the gain setting value is setting information for each frequency band. The gain is, for example, the ratio of the input signal to the output signal expressed in decibels (dB). Each gain setting value is, for example, any integer value from “0” to “9”, and the larger the value, the larger the gain. The gain setting value at which the gain becomes 0 dB is set to an intermediate value (for example, “5”).

Specifically, for example, the noise analysis unit 30 generates a difference signal between the first signal and the second signal, and obtains a frequency region (estimated noise region) of the noise component based on the difference signal. The difference signal includes all or part of the noise component included in the signal output from the detector 20. According to the equation (1), the difference signal is, for example, U (bold body) −L (bold body) = U (bold body) ₀ −L (bold body) ₀ . Then, the noise analysis unit 30 determines the nature of the noise component based on the signals (estimated noise signals) in the estimated noise regions of the first signal and the second signal. For example, when the noise analysis unit 30 detects a signal whose signal intensity is equal to or higher than a predetermined level in a shorter time than a predetermined time determined from the estimated noise signal, the detected signal (detection signal) is a pulse. It is determined that the noise component is sex. In addition, for example, in the case of normal musical sounds, the wideband component is at a lower level than the lowband component, while in the case of pulsed noise, noise analysis is performed using the property that the wideband component is at a higher level than the lowband component. The unit 30 determines that a wide-area component having a level larger than the average level within a predetermined short time is a pulsed noise component. In addition, for example, the noise analysis unit 30 determines that a component having a large autoregressive error (residual signal when the signal is linearly predicted) is a pulse noise component. For example, when the noise analyzing unit 30 detects a signal having a substantially constant signal intensity over a range wider than a predetermined frequency range determined in advance from the estimated noise signal, the detected signal is detected as white noise. It is determined that the noise component is sex.

  When the noise analysis unit 30 determines that the noise component is distributed over a relatively wide frequency range, that is, when it is determined that the signal (detection signal) detected from the estimated noise signal is not pulse noise, the detection signal Determine the gain setting value to reduce the level of. For example, the noise analysis unit 30 reduces the gain for the frequency band corresponding to the estimated noise region in accordance with an arbitrary equalizing operation accompanying subjective evaluation by an operator (evaluator, user, etc.), or the estimated noise region. The gain setting value is determined by increasing the gain for the frequency region that does not fall under. Alternatively, the noise analysis unit 30 determines the gain setting value so as to decrease the level of the detection signal by a certain percentage, for example. In this way, the noise analysis unit 30 performs gains for the in-phase component gain adjustment units 91-1 to 91-n, the first uncorrelated signal gain adjustment unit 92U, and the second uncorrelated signal gain adjustment unit 92L. Setting values are determined, and these gain setting values are stored in the storage unit 40.

  Further, when the noise analysis unit 30 determines that the signal detected from the estimated noise signal is pulse noise, the noise analysis unit 30 determines a filter coefficient for the filter unit 110 for removing the detection signal or reducing the level of the detection signal. To do. For example, the noise analysis unit 30 arbitrarily determines a filter coefficient in accordance with an operation by the operator. Or the noise analysis part 30 determines a filter coefficient based on the estimated noise area | region and the level of a detection signal, for example. The noise analysis unit 30 stores the determined filter coefficient in the storage unit 40.

  The storage unit 40 stores gain setting values for the in-phase component gain adjusting units 91-1 to 91-n, the first uncorrelated signal gain adjusting unit 92U, and the second uncorrelated signal gain adjusting unit 92L. In addition, the storage unit 40 stores filter coefficients for the filter 110. The storage unit 40 is realized by, for example, a semiconductor storage device (for example, a nonvolatile memory).

  The band dividing unit 50 takes in the first signal and the second signal supplied from the detecting unit 20, and divides each of the first signal and the second signal into predetermined bands to generate a plurality of band signals. When the baseband signal is an audio signal, for example, the band dividing unit 50 divides each of the first signal and the second signal for each band having a fixed width to generate a plurality of band signals. Specifically, the band dividing unit 50, for example, the first signal, the first band of 0Hz to less than 1.5KHz, the second band of 1.5KHz to less than 3.0KHz, 3.0KHz to less than 4.5KHz Dividing into a third band, a fourth band of 4.5 KHz or more and less than 6.0 KHz, a fifth band of 6.0 KHz or more and less than 7.5 KHz, and a sixth band of 7.5 KHz or more and less than 9.0 KHz, A first band signal is generated. Similarly, the band dividing unit 50 divides the second signal into the six bands to generate six second band signals. In the following description, the first to sixth bands are defined as 0 Hz to 1.5 KHz, 1.5 KHz to 3.0 KHz, 3.0 KHz to 4.5 KHz, 4.5 KHz to 6.0 KHz, 6.0 KHz to 7.KHz. It is described as 5 KHz and 7.5 KHz to 9.0 KHz. Note that the band dividing unit 50 is not limited to the above example, and may divide the band more finely (increase the number of divisions) or may divide the band with a non-fixed width. The band division method is appropriately determined by design or the like.

  The band dividing unit 50 supplies the first band signal and the second band signal in the same band to the in-phase component extraction unit 60. For example, the first band signal and the second band signal in the first band of 0 Hz to 1.5 KHz are supplied to the in-phase component extraction unit 60-1. In addition, the first band signal and the second band signal in the second band of 1.5 KHz to 3.0 KHz are supplied to the in-phase component extraction unit 60-2. In addition, the first band signal and the second band signal in the third band of 3.0 KHz to 4.5 KHz are supplied to the in-phase component extraction unit 60-3. In addition, the first band signal and the second band signal in the fourth band of 4.5 KHz to 6.0 KHz are supplied to the in-phase component extraction unit 60-4. Further, the first band signal and the second band signal in the fifth band of 6.0 KHz to 7.5 KHz are supplied to the in-phase component extraction unit 60-5. In addition, the first band signal and the second band signal in the sixth band of 7.5 KHz to 9.0 KHz are supplied to the in-phase component extraction unit 60-6. The band dividing unit 50 is realized by a filter bank such as an orthogonal mirror filter, for example.

  The in-phase component extracting unit 60 takes in the first band signal and the second band signal in the same band supplied by the band dividing unit 50, and between the first signal and the second signal from the first band signal and the second band signal. In-phase component signal (in-phase component signal) in the band is extracted. The in-phase component extraction unit 60 performs, for example, adaptive band processing between the first signal and the second signal from the first band signal and the second band signal by executing an adaptive filter process using a learning identification method using an adaptive filter. The in-phase component signal at is extracted. That is, the in-phase component extraction units 60-1 to 60-n extract the in-phase component signal C (bold body) in the equation (1).

  Then, the in-phase component extracting unit 60 supplies the in-phase component signal in the band to the gain adjusting unit 90, the first uncorrelated signal extracting unit 80U, and the second uncorrelated signal extracting unit 80L. For example, the in-phase component extracting unit 60-1 supplies the in-phase component signal in the first band of 0 Hz to 1.5 KHz to the gain adjusting unit 90, the first uncorrelated signal extracting unit 80U, and the second uncorrelated signal extracting unit 80L. To do. Further, the in-phase component extracting unit 60-2 converts the in-phase component signal in the second band of 1.5 KHz to 3.0 KHz into the gain adjusting unit 90, the first uncorrelated signal extracting unit 80U, and the second uncorrelated signal extracting unit 80L. To supply. The in-phase component extraction unit 60-3 converts the in-phase component signal in the third band of 3.0 KHz to 4.5 KHz into the gain adjustment unit 90, the first uncorrelated signal extraction unit 80U, and the second uncorrelated signal extraction unit 80L. To supply. Further, the in-phase component extracting unit 60-4 converts the in-phase component signal in the fourth band of 4.5 KHz to 6.0 KHz into the gain adjusting unit 90, the first uncorrelated signal extracting unit 80U, and the second uncorrelated signal extracting unit 80L. To supply. Further, the in-phase component extracting unit 60-5 converts the in-phase component signal in the fifth band of 6.0 KHz to 7.5 KHz into the gain adjusting unit 90, the first uncorrelated signal extracting unit 80U, and the second uncorrelated signal extracting unit 80L. To supply. Further, the in-phase component extracting unit 60-6 converts the in-phase component signal in the sixth band from 7.5 KHz to 9.0 KHz into the gain adjusting unit 90, the first uncorrelated signal extracting unit 80U, and the second uncorrelated signal extracting unit 80L. To supply. Details of the in-phase component extraction unit 60 will be described later.

  The first signal delay unit 70U receives the first signal supplied from the detection unit 20, and delays the first signal delayed by the total processing time obtained by adding the processing times of the band division unit 50 and the in-phase component extraction unit 60, respectively. Is a delay circuit. Accordingly, the first signal output from the first signal delay unit 70U and the in-phase component signal output from the in-phase component extraction unit 60 are supplied to the first uncorrelated signal extraction unit 80U as a signal at the same time.

  Similarly, the second signal delay unit 70L is a delay circuit that takes in the second signal supplied from the detection unit 20 and outputs a second signal delayed by the same total processing time as described above. Thus, the second signal output from the second signal delay unit 70L and the in-phase component signal output from the in-phase component extraction unit 60 are supplied to the second uncorrelated signal extraction unit 80L as a signal at the same time.

The first uncorrelated signal extraction unit 80U takes in-phase component signals of each band supplied by the in-phase component extraction units 60-1 to 60-n. Also, the first uncorrelated signal extraction unit 80U takes in the first signal delayed by the total processing time supplied by the first signal delay unit 70U. Then, the first uncorrelated signal extraction unit 80U extracts the difference between the first signal and the in-phase component signal over the entire band at the same time as the first uncorrelated signal. Specifically, the first uncorrelated signal extraction unit 80U extracts a signal obtained by subtracting the in-phase component signal over the entire band from the first signal at the same time as the first uncorrelated signal. Thus, the first decorrelated signal extracting section 80U includes a first decorrelated signal U ₀ obtained by subtracting the in-phase component signal C (bold) from the first signal U (bold) shown in the above formula (1) obtain. The first uncorrelated signal extraction unit 80U supplies the first uncorrelated signal to the gain adjustment unit 90.

Also, the second uncorrelated signal extraction unit 80L takes in-phase component signals of each band from the in-phase component extraction units 60-1 to 60-n. The second uncorrelated signal extraction unit 80L takes in the second signal delayed by the total processing time supplied by the second signal delay unit 70L. Then, the second uncorrelated signal extraction unit 80L extracts the difference between the second signal and the in-phase component signal over the entire band at the same time as the second uncorrelated signal. Specifically, the second uncorrelated signal extraction unit 80L extracts a signal obtained by subtracting the in-phase component signal over the entire band from the second signal at the same time as the second uncorrelated signal. Thus, the second decorrelated signal extracting unit 80L includes a second decorrelated signal L ₀ obtained by subtracting the in-phase component signal C (bold) from the second signal L (bold) shown in the above formula (1) obtain. The second uncorrelated signal extraction unit 80L supplies the second uncorrelated signal to the gain adjustment unit 90.

  The control unit 10 includes, for example, a central processing unit (CPU) and a memory that stores a control program that can be executed by the CPU. The CPU functions as the control unit 10 by reading the control program from the memory and executing it. The control unit 10 sets each gain setting value corresponding to each of the in-phase component gain adjustment units 91-1 to 91-n, the first uncorrelated signal gain adjustment unit 92U, and the second uncorrelated signal gain adjustment unit 92L. Read from the storage unit 40. Then, the control unit 10 sets the read gain setting values to the in-phase component gain adjustment units 91-1 to 91-n, the first uncorrelated signal gain adjustment unit 92U, and the second uncorrelated signal gain adjustment unit 92L. And set to In addition, the control unit 10 reads the filter coefficient from the storage unit 40 and sets the read filter coefficient in the filter unit 110.

  The gain adjustment unit 90 takes in the in-phase component signal for each band supplied by each of the in-phase component extraction units 60-1 to 60-n. Further, the gain adjusting unit 90 takes in the first uncorrelated signal supplied from the first uncorrelated signal extracting unit 80U and takes in the second uncorrelated signal supplied from the second uncorrelated signal extracting unit 80L. Then, the gain adjusting unit 90 adjusts the levels of the in-phase component signal, the first uncorrelated signal, and the second uncorrelated signal for each band according to the gain setting value stored in the storage unit 40.

  The in-phase component gain adjustment unit 91 takes in the in-phase component signal in the predetermined band supplied by the in-phase component extraction unit 60 and adjusts the level of the in-phase component signal with the gain set by the control unit 10. For example, the in-phase component gain adjustment unit 91-1 takes in the in-phase component signal in the first band of 0 Hz to 1.5 KHz supplied by the in-phase component extraction unit 60-1, and adjusts the level of the in-phase component signal. The in-phase component gain adjustment unit 91-2 takes in the in-phase component signal in the second band of 1.5 KHz to 3.0 KHz supplied by the in-phase component extraction unit 60-2 and adjusts the level of the in-phase component signal. . Further, the in-phase component gain adjustment unit 91-3 takes in the in-phase component signal in the third band of 3.0 KHz to 4.5 KHz supplied by the in-phase component extraction unit 60-3, and adjusts the level of the in-phase component signal. . Further, the in-phase component gain adjustment unit 91-4 takes in the in-phase component signal in the fourth band of 4.5 KHz to 6.0 KHz supplied by the in-phase component extraction unit 60-4, and adjusts the level of the in-phase component signal. . The in-phase component gain adjusting unit 91-5 takes in the in-phase component signal in the fifth band from 6.0 KHz to 7.5 KHz supplied by the in-phase component extracting unit 60-5 and adjusts the level of the in-phase component signal. . Further, the in-phase component gain adjustment unit 91-6 takes in the in-phase component signal in the sixth band from 7.5 KHz to 9.0 KHz supplied by the in-phase component extraction unit 60-6 and adjusts the level of the in-phase component signal. .

The first uncorrelated signal gain adjustment unit 92U takes in the first uncorrelated signal supplied by the first uncorrelated signal extraction unit 80U, and adjusts the level of the first uncorrelated signal with the gain set by the control unit 10.
The second uncorrelated signal gain adjustment unit 92L takes in the second uncorrelated signal supplied from the second uncorrelated signal extraction unit 80L, and adjusts the level of the second uncorrelated signal with the gain set by the control unit 10.

  The gain adjusting unit 90 supplies the synthesizing unit 100 with the in-phase component signal, the first uncorrelated signal, and the second uncorrelated signal in each band after the level adjustment. For example, the in-phase component gain adjustment unit 91-1 supplies the synthesizing unit 100 with the in-phase component signal after level adjustment in the first band of 0 Hz to 1.5 KHz. Further, the in-phase component gain adjustment unit 91-2 supplies the combining unit 100 with the in-phase component signal after level adjustment in the second band of 1.5 KHz to 3.0 KHz. Further, the in-phase component gain adjustment unit 91-3 supplies the combining unit 100 with the in-phase component signal after level adjustment in the third band of 3.0 KHz to 4.5 KHz. Further, the in-phase component gain adjustment unit 91-4 supplies the synthesizing unit 100 with the in-phase component signal after level adjustment in the fourth band of 4.5 KHz to 6.0 KHz. Further, the in-phase component gain adjusting unit 91-5 supplies the combining unit 100 with the in-phase component signal after level adjustment in the fifth band of 6.0 KHz to 7.5 KHz. Further, the in-phase component gain adjusting unit 91-6 supplies the synthesizing unit 100 with the in-phase component signal after level adjustment in the sixth band of 7.5 KHz to 9.0 KHz. The first decorrelated signal gain adjustment unit 92U supplies the level-adjusted first decorrelated signal to the synthesis unit 100. Further, the second uncorrelated signal gain adjusting unit 92L supplies the second uncorrelated signal after level adjustment to the synthesizing unit 100.

  The synthesizing unit 100 takes in the in-phase component signal, the first uncorrelated signal, and the second uncorrelated signal in each band after the level adjustment supplied from the gain adjusting unit 90. For example, the synthesizing unit 100 captures the in-phase component signal after level adjustment in the first band of 0 Hz to 1.5 KHz supplied by the in-phase component gain adjusting unit 91-1. Further, the synthesizing unit 100 takes in the in-phase component signal after level adjustment in the second band of 1.5 KHz to 3.0 KHz supplied by the in-phase component gain adjusting unit 91-2. The synthesizing unit 100 takes in the in-phase component signal after level adjustment in the third band of 3.0 KHz to 4.5 KHz supplied by the in-phase component gain adjusting unit 91-3. Further, the synthesizing unit 100 takes in the in-phase component signal after level adjustment in the fourth band from 4.5 KHz to 6.0 KHz supplied by the in-phase component gain adjusting unit 91-4. Further, the synthesizing unit 100 captures the in-phase component signal after level adjustment in the fifth band of 6.0 KHz to 7.5 KHz supplied by the in-phase component gain adjusting unit 91-5. Further, the synthesizing unit 100 captures the in-phase component signal after level adjustment in the sixth band of 7.5 KHz to 9.0 KHz supplied by the in-phase component gain adjusting unit 91-6. Further, the synthesizing unit 100 takes in the first uncorrelated signal after level adjustment supplied from the first uncorrelated signal gain adjusting unit 92U. The synthesizing unit 100 captures the level-adjusted second uncorrelated signal supplied from the second uncorrelated signal gain adjusting unit 92L.

  The synthesizing unit 100 adds the in-phase component signal, the first uncorrelated signal, and the second uncorrelated signal in each band after the level adjustment to generate an output signal, and outputs the output signal to the filter unit 110. Supply.

  The filter unit 110 takes in the output signal supplied from the synthesizing unit 100 and performs a filtering process on the output signal based on the filter coefficient stored in the storage unit 40. In other words, the filter unit 110 performs output signal filtering based on the analysis result of the noise component by the noise analysis unit 30. By filtering the output signal by the filter unit 110, it is possible to remove or reduce pulse noise from the output signal.

  FIG. 3 is a block diagram showing a functional configuration of the in-phase component extraction unit 60. As shown in the figure, the in-phase component extraction unit 60 includes a signal acquisition unit 61U, a signal acquisition unit 61L, an adaptive filter unit 62U, an adaptive filter unit 62L, a filter coefficient generation unit 63U, and a filter coefficient generation unit 63L. A delay unit 64U, a delay unit 64L, an error signal detection unit 65U, an error signal detection unit 65L, an in-phase signal generation unit 66, and a signal output unit 67.

The first band signal at time t in the predetermined band supplied by the band dividing unit 50 is U (bold body) (t), and the second band signal is L (bold body) (t).
The signal acquisition unit 61U takes in the first band signal U (bold body) (t) at time t, and uses the first band signal U (bold body) (t) as an adaptive filter unit 62U and a filter coefficient generation unit 63U. To the delay unit 64L.
In addition, the signal acquisition unit 61L takes in the second band signal L (bold body) (t) at time t, and uses the second band signal L (bold body) (t) as an adaptive filter unit 62L and a filter coefficient generation unit. 63L and delay unit 64U.

A filter coefficient generation unit 63U is connected to the adaptive filter unit 62U, and a filter coefficient generation unit 63L is connected to the adaptive filter unit 62L. Adaptive filter unit 62U and filter coefficient generation unit 63U, and adaptive filter unit 62L and filter coefficient generation unit 63L each perform adaptive filter processing to adaptively perform filter coefficient H (bold body) _U and filter coefficient. H (bold body) _L is obtained. Note that, for example, a configuration such as a finite impulse response filter (FIR filter) or an infinite impulse response filter (IIR filter) can be applied to the adaptive filter processing described above. Filter configuration and update algorithm can be selected as appropriate.

A signal obtained by convolution of the filter coefficient H (bold body) _U with the first band signal U (bold body) (t) by the adaptive filter unit 62U is defined as C (bold body) _U (t). A signal obtained by convolution of the filter coefficient H (bold body) _L with the second band signal L (bold body) (t) by the adaptive filter unit 62L is defined as C (bold body) _L (t).

  The delay unit 64U generates a signal L (bold body) '(t) obtained by delaying the second band signal L (bold body) (t) by M / 2 (M is an adaptive filter length, the same applies hereinafter). The delay unit 64L generates a signal U (bold body) '(t) obtained by delaying the first band signal U (bold body) (t) by M / 2.

The adaptive filter unit 62U and the adaptive filter unit 62L include an error signal detection unit 65U and an error signal for calculating an error signal of the signal C (bold body) _U (t) and the signal C (bold body) _L (t), respectively. The signal is supplied to the signal detector 65L. The delay unit 64U and the delay unit 64L supply the signal L (bold body) ′ (t) and the signal U (bold body) ′ (t) to the error signal detection unit 65U and the error signal detection unit 65L, respectively. .

The error signal detection unit 65U receives the signal C (bold body) _U (t) supplied from the adaptive filter unit 62U, and receives the signal L (bold body) ′ (t) supplied from the delay unit 64U. Then, the error signal detector 65U generates an error signal e (bold body) _U (t) obtained by subtracting the signal C (bold body) _U (t) from the signal L (bold body) ′ (t). Further, the error signal detection unit 65L generates an error signal e (bold body) _L (t) obtained by subtracting the signal C (bold body) _L (t) from the signal U (bold body) ′ (t).

The error signal detection unit 65U and the error signal detection unit 65L are configured to generate an error signal e (bold body) _U (t) and an error signal e (bold body) _L (t), respectively, as a filter coefficient generation unit 63U and a filter coefficient generation unit. Feedback to 63L. The filter coefficient generation unit 63U and the adaptive filter unit 62U use the error signal e (bold field) _U (t) to sequentially update the adaptive filter unit 62U using an adaptive algorithm, thereby generating a signal C (bold field) _U (t). Is generated. Further, the filter coefficient generation unit 63L and the adaptive filter unit 62L use the error signal e (bold body) _L (t) to update the adaptive filter unit 62L using an adaptive algorithm, thereby obtaining a signal C (bold body) _L ( t).

The adaptive filter unit 62U and the adaptive filter unit 62L also supply the signal C (bold body) _U (t) and the signal C (bold body) _L (t) to the in-phase signal generation unit 66, respectively. The in-phase signal generation unit 66 takes in the signal C (bold body) _U (t) and the signal C (bold body) _L (t) supplied from the adaptive filter unit 62U and the adaptive filter unit 62L, and outputs the signal C (bold body). ) Add _U (t) and signal C (bold) _L (t). Then, the in-phase signal generation unit 66 supplies the signal output unit 67 with the in-phase component signal C (bold body) ′ (t) obtained by multiplying the addition result by 0.5. That is, the in-phase component signal C (bold body) ′ (t) is (C (bold body) _U (t) + C (bold body) _L (t)) / 2.

  The signal output unit 67 takes in the in-phase component signal C (bold body) '(t) supplied from the in-phase signal generation unit 66 and outputs the in-phase component signal C (bold body)' (t).

The in-phase component extraction unit 60 includes a first band signal U (bold body) (first band signal U (bold body) = C (bold body) + U (bold body) ₀ ) and a signal input to the signal acquisition unit 61U. For the second band signal L (bold body) (second band signal L (bold body) = C (bold body) + L (bold body) ₀ ) input to the acquisition unit 61L, the in-phase component C (bold body) The in-phase component signal C (bold body) is output from the signal output unit 67. The in-phase component extraction unit 60 performs, for example, adaptive filter processing using a learning identification method (Normalized Least Mean Square Algorithm; NLMS algorithm). At this time, the step size parameters are μ = 0.01 and Δ = 0.0001 (= 1 × 10 ⁻⁶ ). Then, the in-phase component extraction unit 60 updates the error signal e (bold body) _U (t) and the error signal e (bold body) _L (t) to be the minimum, thereby obtaining the in-phase component signal C (bold body). Is generated. The filter coefficients H (bold body) _U and H (bold body) _L in the adaptive filter unit 62U and the adaptive filter unit 62L are as shown in the following equations (2) and (3), respectively.

  As described in detail above, the receiving apparatus 1 according to an embodiment of the present invention demodulates the received double sideband amplitude modulation signal and extracts the first signal for the upper sideband and the second signal for the lower sideband. The detection unit 20 is provided. And the receiver 1 divides each of the first signal and the second signal for each band to generate a plurality of band signals, and between the first signal and the second signal from the plurality of band signals. In-phase component extraction units 60-1 to 60-n that extract in-phase component signals for each band. Then, the receiving device 1 extracts the difference between the first signal and each in-phase component signal as a first uncorrelated signal, the first uncorrelated signal extraction unit 80U, and the difference between the second signal and each in-phase component signal. And a second uncorrelated signal extraction unit 80L that extracts two uncorrelated signals. The receiving apparatus 1 includes a gain adjusting unit 90 that adjusts the levels of the in-phase component signals, the first uncorrelated signal, and the second uncorrelated signal, and the in-phase component signals and the first uncorrelated signals after the level adjustment. And a synthesizing unit 90 that adds the second uncorrelated signal and generates an output signal.

By configuring in this way, the receiving apparatus 1 uses the first signal and the second signal obtained by demodulating the double sideband amplitude modulation signal including the noise component mixed in the upper sideband and the lower sideband. The in-phase component signal and the uncorrelated signal (first uncorrelated signal and second uncorrelated signal) for each band are extracted. And the receiver 1 can remove or reduce a noise component by adjusting the addition ratio of the in-phase component signal, the first uncorrelated signal, and the second uncorrelated signal for each band.
Therefore, the receiving apparatus 1 can remove or reduce noise mixed in the double sidebands.

The receiving apparatus 1 according to an embodiment of the present invention further includes a noise analysis unit 30 that analyzes a noise component based on a difference between the first signal and the second signal. The gain adjusting unit 90 adjusts the level based on the analysis result of the noise component by the noise analyzing unit 30. Here, the noise analysis unit 30 determines the nature of the noise component mixed in the double sidebands. For example, the noise analysis unit 30 determines whether the noise component is white noise noise or pulse noise. When the noise component is white noise, the gain adjusting unit 90 adjusts the levels of the in-phase component signal, the first uncorrelated signal, and the second uncorrelated signal with the gain determined for each band.
As described above, the receiving apparatus 1 can effectively remove or reduce the white noise noise component.

In addition, the receiving device 1 according to an embodiment of the present invention includes a storage unit 40 that stores a gain setting value of the gain adjustment unit 90 according to the analysis result of the noise component by the noise analysis unit 30, and a gain from the storage unit 40. A control unit 10 is further provided for reading the set value and setting the gain set value in the gain adjusting unit 90. The gain adjustment unit 90 adjusts the level according to the gain setting value set by the control unit 10.
With this configuration, the receiving device 1 can store a gain setting value set in the gain adjusting unit 90. Therefore, the receiving device 1 can hold the gain setting value determined and evaluated at the time of production, sound quality evaluation, installation, and the like, and can arbitrarily set the gain adjustment unit 90.

The receiving apparatus 1 according to an embodiment of the present invention further includes a filter unit 110 that performs a filtering process on the output signal generated by the synthesis unit 100 based on the analysis result of the noise component by the noise analysis unit 30. When the noise component is pulse noise, the filter unit 110 removes or reduces noise by applying the filter coefficient determined by the noise analysis unit 30.
With this configuration, the receiving device 1 can remove or reduce both white noise noise and pulse noise.

  In the embodiment described above, a plurality of types of gain setting values for the gain adjustment unit 90 may be stored in the storage unit 40. For example, a plurality of types of gain setting values determined by an evaluation experiment or the like of the receiving device 1 are stored in the storage unit 40 in advance. In this case, the receiving device 1 includes an operation unit that receives an operation by an operator. Then, the receiving device 1 controls the control unit 10 according to the operation of the operation unit by the operator, and causes the control unit 10 to read a desired gain setting value from the storage unit 40 and set the gain adjustment unit 90.

  Similarly, a plurality of types of filter coefficients for the filter unit 110 may be stored in the storage unit 40.

  The receiving device 1 may be configured not to include the filter unit 110 and the storage unit 40 or any one of them.

  As mentioned above, although embodiment of this invention was explained in full detail with reference to drawings, the specific structure is not restricted to that embodiment, The design of the range which does not deviate from the summary of this invention, etc. are included.

DESCRIPTION OF SYMBOLS 1 Receiver 10 Control part 20 Detection part 30 Noise analysis part 40 Storage part 50 Band division part 60-1-60-n In-phase component extraction part 61U, 61L Signal acquisition part 62U, 62L Adaptive filter part 63U, 63L Filter coefficient production | generation part 64U, 64L delay unit 65U, 65L error signal detection unit 66 in-phase signal generation unit 67 signal output unit 70U first signal delay unit 70L second signal delay unit 80U first uncorrelated signal extraction unit 80L second uncorrelated signal extraction unit 90 gain adjusting unit 91-1 to 91-n in-phase component gain adjusting unit 92U first uncorrelated signal gain adjusting unit 92L second uncorrelated signal gain adjusting unit 100 combining unit 110 filter unit

Claims (5)

A detector that demodulates the received double sideband amplitude modulation signal and extracts a first signal for the upper sideband and a second signal for the lower sideband;
A band dividing unit that divides each of the first signal and the second signal extracted by the detection unit for each band to generate a plurality of band signals;
An in-phase component extraction unit that extracts a signal of the in-phase component for each of the bands between the first signal and the second signal from the plurality of band signals generated by the band dividing unit;
The difference between the first signal and the signal of each in-phase component extracted by the in-phase component extraction unit is extracted as a first uncorrelated signal, and the difference between the second signal and the signal of each in-phase component is second A non-correlated signal extraction unit that extracts the correlation signal;
A gain adjusting unit for adjusting levels of the signals of the in-phase components and the first uncorrelated signal and the second uncorrelated signal extracted by the uncorrelated signal extracting unit;
A combining unit that generates an output signal by adding the signals of the in-phase components, the first uncorrelated signal, and the second uncorrelated signal, each of which is adjusted by the gain adjusting unit;
A receiving device. The in-phase component extraction unit performs adaptive filter processing based on a learning identification method for each band, so that the in-phase component for each band between the first signal and the second signal is obtained from the plurality of band signals. Extract the signal of the
The receiving apparatus according to claim 1. A noise analysis unit that analyzes a noise component based on a difference between the first signal and the second signal extracted by the detection unit;
The gain adjustment unit adjusts the level based on the analysis result of the noise component by the noise analysis unit,
The receiving apparatus according to claim 1, wherein the receiving apparatus is a receiver. A storage unit that stores a gain setting value of the gain adjustment unit according to the analysis result of the noise component by the noise analysis unit;
A control unit that reads the gain setting value from the storage unit and sets the gain setting value in the gain adjustment unit;
Further comprising
The gain adjustment unit adjusts the level according to the gain setting value set by the control unit,
The receiving device according to claim 3. Based on the analysis result of the noise component by the noise analysis unit, a filter unit that performs a filtering process on the output signal generated by the synthesis unit;
The receiving apparatus according to claim 3, further comprising:

Images